Abstract

COMMENTARIES Starkey Duncan, Jr. University of Chicago action. Conventions provide a framework within which a It is striking that the authors raise the issue of rapport in the context of nonverbal communication research. That such a participant's actions and variations in action acquire mean- complex, high-level issue is addressed at all seems to suggest ing, thus conveying a wide variety of information to the the degree to which the field as a whole has developed. partner, including information relevant to rapport. By the The issues raised in this article were most evocative for same token, conventions segment the "continuous stream of me. Some experiences with the perplexing and utterly unpre- action" into "socially meaningful discrete units," just as the dictable development of rapport-or the failure to develop convention of language segments the continuous speech it-first drew me into studying face-to-face interaction. stream into meaningful, perceptually discrete units. While a graduate student, I worked as a research assistant I view interactions as being constructed through conven- testing applicants for psychotherapy, as part of a larger pro- tions. This would be equally true of interactions involving strangers and of interactions involving family or old friends. ject. The test involved a sort